Zincir, İbrahim
Loading...
Name Variants
Job Title
Dr.Öğr.Üyesi
Email Address
Main Affiliation
01.01.09.01. Bilgisayar Mühendisliği Bölümü
Status
Former Staff
Website
ORCID ID
Scopus Author ID
Turkish CoHE Profile ID
Google Scholar ID
WoS Researcher ID
Sustainable Development Goals
SDG data is not available
This researcher does not have a Scopus ID.
Documents
13
Citations
239
Scholarly Output
9
Articles
0
Views / Downloads
0/0
Supervised MSc Theses
9
Supervised PhD Theses
0
WoS Citation Count
0
Scopus Citation Count
0
Patents
0
Projects
0
WoS Citations per Publication
0.00
Scopus Citations per Publication
0.00
Open Access Source
0
Supervised Theses
9
Journals data is not available
Scopus Quartile Distribution
Quartile distribution chart data is not available
Competency Cloud
9 results
Scholarly Output Search Results
Now showing 1 - 9 of 9
Master Thesis Veri madenciliği ile üretim firesini tahminleme(2018) Karadağ, Gözde; Zincir, İbrahimEfficiency is the most important criteria in production systems. The more wastage rate means the less efficiency in the production. However, companies accept wastage to some extent due to dynamism of production. There are two wastage types in production such as setup wastage and production wastage. Setup wastage is not changeable without changing resources. It occurs while preparing materials and machines to the production. This part forms fifteen percent of the total wastage in this study. Therefore, the acceptable wastage is fifteen percent. This research aims to estimate the wastage rate of a packaging company by using data mining methods. In addition, the wastage rate range is decreased gradually in order to make more accurate estimations. The predicted result is specified as yes, no. If the wastage range is less than 15%, the prediction is no, otherwise yes. In the second analysis, the range is divided to three categories to make prediction more specific which are S, M, L. If the wastage rate is less than 14% it is defined as S. If the wastage is between 14% and 16%, the prediction is M and if it is more than 16%, then it is L. In order to investigate the effect of the parameters in production, 10 versions are generated for each group. Each version has different production parameters. Twenty algorithms are applied these twenty data sets. As a result of the study, either the range limit is taken as 15% or in the range of 14% and 16% same results are obtained. Naïve Bayes algorithm with version 4 has the best result for Group1 and Group2. It shows that the wastage is not related with month and customer information as much as other parameters. Key Words: production systems, data mining, wastage, production attributesMaster Thesis Portatif EEG cihazıyla beyin dalgalarının mobil platformlarda analiz edilmesi(2016) Kalkavan, Tuğçe; Zincir, İbrahimElektroensefalografi kısaca EEG, beynin elektriksel hareketlerini ölçen bilişsel bilim, sinirbilim ve psikofizyolojik araştırma alanlarında sıkça kullanılan bir alettir. Klinik uygulamalarda kafa derisine değen birden fazla elekrot ile beynin farklı bölgelerinden sinyaller toplanır. Teknolojinin gelişmesi ile birlikte son yıllarda kolay ve uygulanabilir olması açısından portatif EEG cihazları geliştirmek için yatırımlar yapılmaktadır. Yapılan çalışmalarda portatif EEG cihazları hafızayı geliştirmek, beyni görüntüleme ve nöroterapi gibi birçok alanda kullanılmakta. Portatif EEG cihazları cihazları yönetebilmek adına ucuz ve kolay bir çözüm sağlamaktadır. Bu tez, portatif EEG cihazı kullanarak hem donanıma dayalı hem de yazılım bileşenlerini kullanabilmek adına çözüm sunmakta olup, bu alanda yapılmış çalışmaları ve sonuçlarını alt bölümlerde incelemektedir.Master Thesis İnsansız hava araçlarının uzaktan kontrolü için ses tanıma algoritması implementasyonu(2017) Sofuoğlu, İlhan; Zincir, İbrahimThis study aims to develop a system that directs the Bebop Drone (Unmanned Aerial Vehicle)using Turkish isolated voice commands, isolated words of a voice recognition algorithm. Some of the algorithms and methods used in the previous studies have been examined. Some algorithms such as Gaussian Mixture Models, Hidden Markov Models, Dynamic Time Warping, Artificial Neural Networks and Google's Deep Learning Algorithm have been examined in the literature with the studies mentioned as isolated word recognition. In addition, preliminary processing such as sampling, windowing, framing, noise reduction and subtraction of audio features before using these classifier algorithms has been examined, and the effects of these processes on the isolated word recognition process have been compared.Master Thesis Artırılmış gerçeklik kullanan ev dekorasyon uygulaması(2013) Doğan, Uğur Çağrı; Ak, Vedat Can; Karabulut, Korhan; Zincir, İbrahimArtırılmış Gerçeklik, bulunduğumuz gerçek ortamı bilgisayar yardımı ile oluşturulmuş sanal nesneler ile birleştiren bir sistemdir. Sanal Gerçeklik, nesneleri tamamen sanal bir ortamda sergilediğinden, gerçek ortamda sanal nesneleri sergileyebilmek için Artırılmış Gerçeklik daha etkilidir. Bu tez, 3 boyutlu mobilya modellerini kullanıcının bulunduğu gerçek ortamda (özellikle kapalı ortamlarda), dekor değişikliklerini ön-izleme amacıyla görüntüleyen bir Artırılmış Gerçeklik uygulaması sunmaktadır. Tezin amacı, kullanıcılara, bulundukları gerçek ortamda 3 boyutlu mobilya modellerini ve ev tasarım eşyalarını göz önüne getirme imkânını verip, satın alma sürecinde kolaylık sağlamaktır. Bu uygulama, 3 boyutlu modellerin görselleştirmesi için üç teknik kullanmaktadır. İlki, modeli sahneye yerleştirebilmek için işaretleyici takibi yapan, işaretleyici tabanlı bir tekniktir. İkinci teknik de birincisi gibi işaretleyici tabanlı bir tekniktir; ancak birden fazla işaretleyiciyi takip eder. Her bir işaretleyici farklı mobilya modellerine atanmıştır. Ve sonuncusu, gerçek ortamdaki herhangi bir nesneyi, kullanıcıların takip işlemi için kullanabilmesini sağlayan işaretleyicisiz bir tekniktir. Ayrıca, bu tez SIFT ve eş düzlemli POSIT algoritmalarının sonuçlarını birleştiren bir bakış açısı sağlamaktadır. Önerilen sistem farklı ışıklandırma durumlarında başarıyla test edilmiştir.Master Thesis El hareketlerini tanıma için makine öğrenmesi algoritmalarının uygulanması(2019) Keçeci, Aybüke; Zincir, İbrahimHareket tanıma, insan-bilgisayar etkileşimi (HCI) için son derece önemlidir. Bir el hareketi tanıma sistemi, sözlü olmayan iletişimin doğal, yenilikçi ve modern bir yolunu sağlar. İnsan-bilgisayar etkileşimlerinde geniş bir uygulama alanına sahiptir. El hareketlerinin bilgisayarla tanınması, tıbbi sistemler, insan-bilgisayar etkileşimi gibi birçok uygulamada yaygın olarak kullanılmaktadır, çünkü el hareketi tanıma, insanlara doğal ve sezgisel bir bilgisayar ara yüzü sağlar. Bu çalışmada, kullanıcının artık kas hareketlerini; el protezinin açık / kapalı el, el bileğini döndürmek gibi belirli hareketlerini haritalanması amaçlanmıştır. Bu problemleri çözmek için öncelikle, hangi özelliklerin gerekli olduğuna karar vermek için, bazı sütunları varyasyonları halinde çıkararak deneyler gerçekleştirildi. Oluşturulan veri kümesi ile yapay sinir ağları algoritmalarından birçoğu ile deneyler yapılmış, Naive Bayes, BayesNet, Multilayer Perceptron, Bagging, Hoeffding Tree and Random Forest olan en başarılı algoritmalar arasından Random Forest seçilmiştir.Master Thesis Eeg tabanlı robotik kol kontrolü için makine öğrenme algoritmalarının uygulanması(2019) Ayluçtarhan, Gülşen; Zincir, İbrahimElectroencephalography (EEG) analysis has been an important subject of several studies like neuroscience, medical diagnosis and rehabilitation engineering. EEG is widely used with brain-computer interface (BCI) systems because of its ability to use brain signals not muscles to control an external BCI prosthetic device. With the development of technology, it became possible to use large EEG datasets and BCI method to extract an understandable information. In this present work, EEG-based BCI system is used by making participants perform a series of grasping and lifting hand movements. Dataset which consists of EEG and EMG information has been implemented via 15 machine learning algorithms as multiclass classification. The best results came from IB1 algorithm. But, random forest, bagging and classification via regression algorithms also have promising outcomes. Hence, this study successfully proved that it is possible to help patients with no hand function to gain control.Master Thesis Optimization of Convolutional Neural Networks via Graphic Cards for Centralized Data(2019) Cibil, Erinç; Zincir, İbrahimIn this thesis, it is aimed to design a new approach optimized for systems that use multiple graphics processing units (GPU) in order to find highly discriminative attributes of digitized handwritten numbers obtained from MNIST dataset and their results. In this study, the convolutional neural network (CNN) method and digitized handwriting classification method are discussed in three sections. In the first part, the classification is obtained by implanting the naive convolutional neural network into the graphic processing unit. In the second stage, the process layers for graphic processing units are parallelized and the data is adjusted for parallel processing layers and the classification is aimed with optimized memory access pattern approach. In the last stage, the method has been improved to work on more than one graphic processing unit. The aim of this stage is to improve the education time of convolutional neural network inversely proportional to the number of graphic processing units used.Master Thesis Yaşar Üniversitesi uzaktan eğitim sistemi analizi(2016) Kınay, Erhan; Zincir, İbrahimData mining is the computational process of analyzing large data sets, discovering the patterns of data groups and reaching the targeted data through processing operation. The process of extraction of the important information through the large amount of data and execution of that data is called 'Data Mining'. In this research, our aim is to predict a student's grade via implementing machine learning techniques of data mining over Yasar University UFND Dataset. The dataset in question is collected between 2012-2013 Fall Term and 2015-2016 Fall Term and made from the log records of website activity of 10 courses over 5 terms (Yasar University e-learning website https://e.yasar.edu.tr). These logs inherit end user information regarding how and when the website in question is used and studied. By implementing these data mining techniques, first the proposed framework analyzed the collected data and then tried to successfully guess whether a student will pass or fail from the course in question. In order to achieve this goal, the proposed framework trained the system with 31 different classification algorithms and then a final algorithm was selected for each course, for each term and for the combination of the data of each course over 5 terms. These algorithms are Naive Bayes, Ripper, J48, Bayes Net, Adaboost, AdTree, Attribute Selected Classifier, Bagging, Classification via Regression, Conjunctives Rules, CV Parameter Selection, Decision Table, DTNB, END, Filtered Classifier, Grading, IB1, Ibk, Kstar, Logistic, LogiBoost, LWL, MultiBoostAB, Multiclass Classifier, Multi Scheme, Multilayer Perceptron, SMO, Voted Perceptron, Random Forest, and ZeroR. At the end, all these results are analyzed and then evaluated to achieve the goal of effective prediction of a student's success.Master Thesis Siber saldırı tespit etme ve önleme yazılım uygulaması(2017) Kılıç, Mert Can; Zincir, İbrahimComputer networks and computational communication technologies have been improving very fast since the first connection was established between two computers by ARPANET in 1969. The daily routines are becoming digitalized day by day. This transformation provides easiness, but at the same time it causes some security problems. The security mechanisms such as authentication, authorization and recognition that a human brain can automatically execute, can be manipulated in digital environments. The people who have the motivation for stealing information, profiting in illegal ways, blackmailing and so on, use a lot of manipulative methods by making use of computer networks and the systems that are based on these networks. These methods are changing and being updated very rapidly, so it is very difficult to detect and prevent that kind of attacks. Even the new generation tools that have current electronic control mechanisms can be exposed to that kind of attacks, so that it is known that this may cause crucial destructions including death.the security experts who provide service for defending systems against these complicated and sophisticated attacks, may be unaware and uninformed about the security flaws that are being used by the people who have the criminal motivations. The penetration tests that are being conducted periodically, are mostly for the revealed security flaws. Namely, the security flows are updated more frequently than the penetration tests.The systems that are not maintained or operated by the qualified security experts are very open to the old-fashioned attacks, and these poorly maintained systems are avoiding the costs of the sophisticated detection and prevention software.The main goal of this work is to use a x86 based embedded system which hosts a customized Linux based operating system with the dynamic analysis of the both remotely and locally gathered/enumerated logs as well as implementing network security functionalities of the conventional network equipment provide. Thus allowing to gather and analyze information about the local or remote network resulting automated reporting for the IT administrators.
